» Download Article

Authors
Gustaf Hendeby; Fredrik Gustafsson

Digital Object Identifier (DOI)
10.3182/20050703-6-CZ-1902.00046

Page Numbers:
45-45

Index Terms
Kalman filters,linear filters,Cramér-Rao lower bound,nonlinear filters,optimal filtering

Abstract
The Kalman filter is known to be the optimal linear filter for linear non-Gaussian systems. However, nonlinear filters such as Kalman filter banks and more recent numerical methods such as the particle filter are sometimes superior in performance. Here a procedure to a priori decide how much can be gained using nonlinear filters, without having to resort to Monte Carlo simulations, is outlined. The procedure is derived in terms of the posterior Cramér-Rao lower bound. Results are shown for a class of standard distributions and models in practice.

References

[1] Anderson, B. D. O. and J. B. Moore (1979). Optimal
    Filtering. Prentice Hall. Englewood Cliffs, NJ.
[2] Bergman, N. (1999). Recursive Bayesian Estimation:
    Navigation and Tracking Applications. Dissertation
     no. 579. Linköpings universitet.
[3] Bergman, N. (2001). Posterior Cramér-Rao bounds for
    sequential estimation. Chap. 15, pp. 321-338. In:
    Doucet et al. (2001).
[4] Bergman, N., L. Ljung and F. Gustafsson (1999). Terrain
     navigation using Bayesian statistics. IEEE
    Control Syst. Mag. 19(3), 33-40.
[5] Blom, H. A. P. and Y. Bar-Shalom (1988). The interacting
     multiple model algorithm for systems with
    Markovian switching coefficients. IEEE Trans.
    Automat. Contr. 33(8), 780-783.
[6] Cox, D. R. and D. V. Hinkley (1974). Theoretical
    Statistics. Chapman and Hall. New York.
[7] Dahlgren, C. (1998). Nonlinear black box modelling
     of JAS 39 Gripens radar altimeter. Master's
     thesis no LiTH-ISY-EX-1958. Dept. E.E.,
    Linköpings universitet, Sweden.
[8] Doucet, A., de Freitas, N. and Gordon, N. (Eds.)
    (2001). Sequential Monte Carlo Methods in
    Practice. Statistics for Engineering and Information
     Science. Springer Verlag. New York.
[9] Gordon, N. J., D. J. Salmond and A. F. M.
    Smith (1993). Novel approach to nonlinear/
    non-Gausian Bayesian state estimation. IEE
    Proceedings-F 140(2), 107-113.
[10] Gustafsson, F. (2000). Adaptive Filtering and Change
     Detection. John Wiley & Sons, Ltd. Chichester,
     West Sussex, England.
[11] Hendeby, G. and F. Gustafsson (2005). Fundamental
      filtering limitations in linear non-Gaussian
     systems. Technical Report LiTH-ISY-R-2681.
     Dept. E.E., Linköpings universitet, Sweden.
[12] Kailath, T. (1980). Linear Systems. Prentice-Hall, Inc.
[13] Kailath, T., A. H. Sayed and B. Hassibi (2000). Linear
     Estimation. Prentice-Hall, Inc.
[14] Kalman, R. E. (1960). A new approach to linear filtering
      and prediction problems. Transactions of the
     ASME-Journal of Basic Engineering 82 (Series
     D), 35-45.
[15] Kay, S. M. (1993). Fundamentals of Statistical Signal
     Processing: Estimation Theory. Vol. 1. Prentice
     Hall, Inc.
[16] Kay, S. M. (1998). Fundamentals of Statistical Signal
     Processing: Detection Theory. Vol. 2. Prentice
     Hall, Inc.
[17] Kay, S. M. and D. Sengupta (1987). Optimal detection
     in colored non-Gaussian noise with unknown parameter.
      In: IEEE ICASSP '87. Vol. 12. pp. 1087-
     1089.
[18] Robert, C. P. and G. Casella (1999). Monte Carlo
     Statistical Methods. Springer Texts in Statistics.
     Springer-Verlag. New York.
[19] Sorenson, H. W. and D. L. Alspach (1971). Recursive
     Bayesian estimation using Gaussian sums. Automatica
      7(4), 465-479.
[20] Tichavský, P., C. H. Muravchik and A. Nehorai
     (1998). Posterior Cramér-Rao discrete-time non-linear
      filtering. IEEE Trans. Signal Processing
     46(5), 1386-1396.